Air Water (Japan)

Regenerating the true pulp-dentin complex in mature teeth with necrotic pulps or apical lesions remains challenging, and consistent clinical evidence of dentin bridge formation is limited. This report describes two adult cases treated with regenerative endodontic therapy (RET) using autologous dental pulp stem cells (DPSCs) and demineralized dentin matrix (DDM). In Case 1, a 33-year-old woman developed pulp necrosis with apical periodontitis in a maxillary left central incisor following traumatic subluxation. After canal disinfection, autologous DPSCs isolated from her extracted third molar were transplanted into the canal, and DDM particles were placed at the orifice as a capping scaffold. The contralateral traumatized but still vital incisor was monitored as a control. During 52 weeks of follow-up, the treated tooth regained positive pulp sensibility within 1 week, developed a dentin-like bridge exceeding 3mm at the orifice, and on 52-week MRI showed high-signal pulp-like tissue comparable to adjacent healthy teeth, whereas the untreated control lost vitality and underwent progressive pulp canal obliteration. In Case 2, a 48-year-old man presented with a maxillary left central incisor that had previously undergone root canal treatment and exhibited severe coronal dentin loss beneath an old crown. Orthodontic extrusion was performed to expose sound supragingival dentin, followed by RET with autologous DPSCs and DDM. Sensibility testing was not feasible due to coronal coverage; however, at 48 weeks CBCT demonstrated a well-defined dentin-like bridge extending from beneath the canal orifice into the coronal pulp chamber. In both cases, quantitative CBCT showed progressive increases in mineral density at the regenerated tissue, reaching approximately 75-80% of adjacent native dentin by the final follow-up. No adverse events were observed, and there were no signs of ankylosis, root resorption, or generalized canal obliteration. These cases suggest that autologous DPSCs combined with DDM can support dentin-like bridge formation and functional recovery in mature incisors, offering a potential biologically based alternative or adjunct to conventional root canal therapy for preserving structurally compromised teeth.

We have developed a prototype of a needle-sized rigid endoscope using graded-index plastic optical fiber (GI-POF) technology. With highly accurate control of refractive index within a GI-POF, we can utilize it for an ultra-fine lens as a replacement for a relay lens in rigid endoscope. The rigid endoscope with GI-POF lens has a possibility to allow doctors to directly observe a patient's inner affected area before and after surgery in a minimally invasive procedure, enabling rapid and accurate assessment of the patient's condition and efficient post-operative follow-up.

Frequency-domain probe beam deflection (FD-PBD) is an experimental technique for measuring thermal properties that combines heating by a modulated pump laser and measurement of the temperature field via thermoelastic displacement of the sample surface. In the conventional implementation of FD-PBD, the data are mostly sensitive to the in-plane thermal diffusivity. We describe an extension of FD-PBD that introduces sensitivity to through-plane thermal conductance by immersing the sample in a dielectric liquid and measuring the beam deflection created by the temperature field of the liquid. We demonstrate the accuracy of the method by measuring (1) the thermal conductivity of a 310 nm thick thermally grown oxide on Si, (2) the thermal boundary conductance of bonded interface between a 3C-SiC film and a single crystal diamond substrate, and (3) the thermal conductivities of several bulk materials. We map the thermal boundary conductance of a 3C-SiC/diamond interface with a precision of 1% using a lock-in time constant of 3 ms and dwell time of 15 ms. The spatial resolution and maximum probing depth are proportional to the radius of the focused laser beams and can be varied over the range of 1–20 μm and 4–80 μm, respectively, by varying the 1/e2 intensity radius of the focused laser beams from 2 to 40 μm. FD-PBD with liquid immersion thus enables fast mapping of spatial variations in thermal boundary conductance of deeply buried interfaces.

Hydrogen market is expected to glow exponentially and in attracting attention as clean energy source that does not emit CO2. The hydrogen production method by the reaction of water and aluminum powder has been studied. In this study, experiments were conducted on waste aluminum alloy powder discharged from machining factory. The main objective of this study is to investigate the reactivity of waste aluminum alloy powder under high pressure conditions. Initial pressurization of 0, 0.15, 0.3MPaG was applied to the reaction tank to investigate the effect of pressure level on the rate of hydrogen production. The results showed that the chemical reaction was not affected by the application of initial pressurization to the reactor. Furthermore, result of composition analysis of the generated gas confirmed that 98.5wt% of hydrogen was generated even when waste aluminum alloy powder was used.

Thermal management is critical in contemporary electronic systems, and integrating diamondwith semiconductors offers the most promising solution to improve heat dissipation. However, developing a technique that can fully exploit the high thermal conductivity of diamond, withstand high-temperature annealing processes, and enable mass production is a significant challenge. In this study, the successful transfer of AlGaN/GaN/3C-SiC layers grown on Si to a large-size diamond substrate is demonstrated, followed by the fabrication of GaN high electron mobility transistors (HEMTs) on the diamond. Notably, no exfoliation of 3C-SiC/diamond bonding interfaces is observed even after annealing at 1100°C, which is essential for high-quality GaN crystal growth on the diamond. The thermal boundary conductance of the 3C-SiC-diamond interface reaches ≈55MWm-2K-1, which is efficient for device cooling. GaN HEMTs fabricated on the diamond substrate exhibit the highest maximum drain current and the lowest surface temperature compared to those on Si and SiC substrates. Furthermore, the device thermal resistance of GaN HEMTs on the diamond substrate is significantly reduced compared to those on SiC substrates. These results indicate that the GaN/3C-SiC on diamond technique has the potential to revolutionize the development of power and radio-frequency electronics with improved thermal management capabilities.

Currently, the need for telemedicine is increasing all over the world. At the same time, the resolution of various medical images has become higher than before, and resolution of surgical endoscopes in particular have reached 8K Ultra-High-Definition (UHD). We conducted demonstration experiments of 8K-UHD medical image remote transmission with low-latency encoding and decoding between two sites (Tokushima Prefectural Central Hospital and Tokushima Prefectural Miyoshi Hospital) 68km apart. The two sites were connected by a 10-Gbps cable TV network, and on the transmitting side (Tokushima Prefectural Miyoshi Hospital), 8K-UHD endoscope image was compressed from 48 Gbps to 300 Mbps for the local cable TV network by ELL 8K HEVC Encoder. On the receiving side, the compressed image was decoded in real time into 8K-UHD uncompressed 48-Gbps image by ELL 8K HEVC Decoder and displayed on a 70-inch 8K-UHD monitor. The latency between the two sites (from 8K-UHD camera to 8K-UHD monitor) was shortened to about 0.15 second, and we and hospital doctors were able to have a two-way conversation on the 4K video conference system (Vidyo) while watching 8K-UHD medical image without any discomfort. In the future, we will improve latency not only in 8K-UHD encoders/decoders but also in peripheral systems such as various interface conversions on the system and monitors.

Abstract Graphs help to define the relationships between entities in the data. These relationships, represented by edges, often provide additional context information which can be utilised to discover patterns in the data. Graph Neural Networks (GNNs) employ the inductive bias of the graph structure to learn and predict on various tasks. The primary operation of graph neural networks is the feature aggregation step performed over neighbours of the node based on the structure of the graph. In addition to its own features, for each hop, the node gets additional combined features from its neighbours. These aggregated features help define the similarity or dissimilarity of the nodes with respect to the labels and are useful for tasks like node classification. However, in real‐world data, features of neighbours at different hops may not correlate with the node's features. Thus, any indiscriminate feature aggregation by GNN might cause the addition of noisy features leading to degradation in model's performance. In this work, we show that selective aggregation of node features from various hops leads to better performance than default aggregation on the node classification task. Furthermore, we propose a Dual‐Net GNN architecture with a classifier model and a selector model. The classifier model trains over a subset of input node features to predict node labels while the selector model learns to provide optimal input subset to the classifier for the best performance. These two models are trained jointly to learn the best subset of features that give higher accuracy in node label predictions. With extensive experiments, we show that our proposed model outperforms both feature selection methods and state‐of‐the‐art GNN models with remarkable improvements up to 27.8%.

We successfully fabricate a Ga <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> O <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</inf> /3C-SiC direct bonding by the surface-activated bonding (SAB) of 3C-SiC epitaxial layer grown on Si substrate to Ga <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> O <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</inf> substrate. The structure of bonding interface will be investigated by transmission electron microscope (TEM).